检索结果-内蒙古大学图书馆

Disentangled graph Contrastive Learning With Independence Promotion

IEEE TRANSACTIONS ON KNOWLEDGE AND data ENGINEERING 2023年第8期35卷 7856-7869页

作者： Li, Haoyang Zhang, Ziwei Wang, Xin Zhu, Wenwu Tsinghua Univ Dept Comp Sci & Technol Beijing 100084 Peoples R China

Self-supervised learning for graph neural networks has attracted considerable attention and shows notable successes in graph representation learning. However, the formation of a real-world graph typically arises from highly complex interactions of many latent factors. The existing self-supervised learning methods for GNNs are inherently holistic and neglect the entanglement of the latent factors, resulting in suboptimal learned representations for downstream tasks and difficult to be interpreted. Learning disentangled graph representations with self-supervised learning poses great challenges and remains largely ignored by the existing literature. In this paper, we introduce Independence Promoted Disentangled graph Contrastive Learning (IDGCL) method, which can learn disentangled graph-level representations with self-supervision. In particular, we first identify the latent factors of the input graph and derive its factorized representations. Then we propose a factor-wise discrimination objective in a contrastive learning manner, which can force the factorized representations to independently reflect the expressive information from different latent factors. To further promote the independence between the representations, we employ the Hilbert-Schmidt Independence Criterion to eliminate the dependence among different representations, which is effectively integrated into the self-supervised framework as a regularizer. Extensive experiments on synthetic and real-world datasets demonstrate the superiority of our method against several state-of-the-art baselines.

关键词： graph data mining graph neural network self-supervised learning disentangled representation learning

来源：评论

学校读者我要写书评

暂无评论

Propagation Enhanced Neural Message Passing for graph Representation Learning

引用

IEEE TRANSACTIONS ON KNOWLEDGE AND data ENGINEERING 2023年第2期35卷 1952-1964页

作者： Fan, Xiaolong Gong, Maoguo Wu, Yue Qin, A. K. Xie, Yu Xidian Univ Sch Elect Engn Key Lab Intelligent Percept & Image Understanding Minist Educ Xian 710126 Shaanxi Peoples R China Xidian Univ Sch Comp Sci & Technol Xian 710126 Shaanxi Peoples R China Swinburne Univ Technol Dept Comp Technol Melbourne Vic 3122 Australia Shanxi Univ Key Lab Computat Intelligence & Chinese Informat P Minist Educ Taiyuan 030006 Peoples R China

graph Neural Network (GNN) is capable of applying deep neural networks to graph domains. Recently, Message Passing Neural Networks (MPNNs) have been proposed to generalize several existing graph neural networks into a unified framework. For graph representation learning, MPNNs first generate discriminative node representations using the message passing function and then read from the node representation space to generate a graph representation using the readout function. In this paper, we analyze the representation capacity of the MPNNs for aggregating graph information and observe that the existing approaches ignore the self-loop for graph representation learning, leading to limited representation capacity. To alleviate this issue, we introduce a simple yet effective propagation enhanced extension, Self-Connected Neural Message Passing (SC-NMP), which aggregates the node representations of the current step and the graph representation of the previous step. To further improve the information flow, we also propose a Densely Self-Connected Neural Message Passing (DSC-NMP) that connects each layer to every other layer in a feed-forward fashion. Both proposed architectures are applied at each layer and the graph representation can then be used as input into all subsequent layers. Remarkably, combining these two architectures with existing GNN variants can improve these models' performance for graph representation learning. Extensive experiments on various benchmark datasets strongly demonstrate the effectiveness, leading to superior performance for graph classification and regression tasks.

关键词： Message passing Aggregates Task analysis data models Predictive models graph neural networks Adaptation models graph data mining graph representation learning graph neural network neural message passing

来源：评论

学校读者我要写书评

暂无评论

Attacking the Core Structure of Complex Network

引用

IEEE TRANSACTIONS ON COMPUTATIONAL SOCIAL SYSTEMS 2023年第4期10卷 1428-1442页

作者： Zhou, Bo Lv, Yuqian Wang, Jinhuan Zhang, Jian Xuan, Qi Zhejiang Univ Technol Inst Cyberspace Secur Coll Informat Engn Hangzhou 310023 Peoples R China Zhejiang Inst Commun Dept Intelligent Control Hangzhou 311112 Peoples R China PCL Res Ctr Networks & Commun Peng Cheng Lab Shenzhen 518000 Peoples R China Utron Technol Co Ltd Hangzhou 310056 Peoples R China

The concept of k-core in complex networks plays a key role in many applications, e.g., understanding the global structure or identifying central/critical nodes, of a network. A malicious attacker with a jamming ability can exploit the vulnerability of the k-core structure to attack the network and invalidate the network analysis methods, e.g., reducing the k-shell values of nodes can deceive graph algorithms, leading to the wrong decisions. In this article, we investigate the robustness of the k-core structure under adversarial attacks by deleting edges, for the first time. First, we give the general definition of the targeted k-core attack, map it to the set cover problem, which is NP-hard, and further introduce a series of evaluation metrics to measure the performance of attack methods. Then, we propose the Q index theoretically as the probability that the terminal node of an edge does not belong to the innermost core, which is further used to guide the design of our heuristic attack methods, namely, COREATTACK and GreedyCOREATTACK. The experiments on a variety of real-world networks demonstrate that our methods behave much better than a series of baselines, in terms of much smaller edge change rate (ECR) and false attack rate (FAR), achieving state-of-the-art attack performance. More impressively, for certain real-world networks, only deleting one edge from the k-core may lead to the collapse of the innermost core, even if this core contains dozens of nodes. Such a phenomenon indicates that the k-core structure could be extremely vulnerable under adversarial attacks, and its robustness, thus. should be carefully addressed to ensure the security of marie graph algorithms. An open-source implementation is available at https://***/Yocenly/COREATTACCK.

关键词： Adversarial attack AI security graph data mining k-core decomposition network science social network structural robustness

来源：评论

学校读者我要写书评

暂无评论

M2Mol: Multi-view Multi-granularity Molecular Representation Learning for Property Prediction 29th

M2Mol: Multi-view Multi-granularity Molecular Representation...

引用

29th International Conference on database Systems for Advanced Applications (DASFAA)

作者： Zhang, Ran Wang, Xuezhi Liu, Kunpeng Zhou, Yuanchun Wang, Pengfei Chinese Acad Sci Comp Network Informat Ctr Beijing Peoples R China Univ Chinese Acad Sci Beijing Peoples R China Univ Chinese Acad Sci Hangzhou Inst Adv Study Beijing Peoples R China Portland State Univ Portland OR 97207 USA

ISBN: (纸本)9789819755745;9789819755752

Molecular property prediction has drawn considerable attention in drug discovery, material design, and biological system analysis in recent years. Particularly, artificial intelligence-driven computational methods hold great promise in expediting the molecular prediction process. While sequence-based methods and graph-based methods have significantly enhanced molecular property prediction, they still exhibit limitations in capturing intricate multi-dimensional molecular information and fully exploiting multi-level structural units. We present a Multi-view Multi-granularity Molecular representation learning framework, M2Mol. Specifically, M2Mol jointly models the molecular sequential arrangement and structural topology with the sequence view and graph view, respectively. In each view, M2Mol captures atom-level attributes and motif-level semantics to intensify the perception of the molecular multi-granularity structural units. In addition, we design inconsistency loss to promote the alignment between the sequence and graph views, and mutual information loss to capture the complementarity between the atom and motif levels. Finally, extensive experiments conducted on six real-world datasets demonstrate the superior effectiveness of the proposed model across both classification and regression tasks.

关键词： molecular property prediction graph data mining graph neural networks

来源：评论

学校读者我要写书评

暂无评论

Enhancing Multi-view Contrastive Learning for graph Anomaly Detection 29th

Enhancing Multi-view Contrastive Learning for Graph Anomaly ...

引用

29th International Conference on database Systems for Advanced Applications (DASFAA)

作者： Lu, Qingcheng Wu, Nannan Zhao, Yiming Wang, Wenjun Zu, Quannan Tianjin Univ Coll Intelligence & Comp Tianjin Peoples R China Tianjin Univ Dept Management & Econ Tianjin Peoples R China

ISBN: (纸本)9789819755714;9789819755721

graph anomaly detection (GAD) has garnered considerable attention owing to its wide-ranging applications in real-world scenarios, including financial network and cybersecurity. Due to the difficulty associated with learning intricate connectivity and high-dimensional attributes in graphs, GAD task poses significant challenges. Existing deep learning methods extract local information between the target node and its neighbors for GAD. Nevertheless, they neglect the global information in the graph which results in suboptimal performance. In this paper, we propose a self-supervised learning framework integrating contrastive learning module and attribute reconstruction module, which can effectively capture both local and global information for GAD. Specifically, we regard original graph as local view, and global view is generated based on this local view. In contrastive learning module, target node and subgraph are sampled from both views to construct multi-scale contrast combinations. With intra- and intre- view contrastive strategy, we can effectively capture both local and global information of anomaly nodes. In attribute reconstruction module, masked graph auto-encoder is employed to further capture anomaly information within the attribute space. Extensive experiments are conducted on six benchmark datasets and results reveal that our approach outperforms the current state-of-the-art methods.

关键词： graph data mining graph Anomaly Detection Contrastive Learning

来源：评论

学校读者我要写书评

暂无评论

Deep Neural Message Passing With Hierarchical Layer Aggregation and Neighbor Normalization

引用

IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS 2022年第12期33卷 7172-7184页

作者： Fan, Xiaolong Gong, Maoguo Tang, Zedong Wu, Yue Xidian Univ Sch Elect Engn Key Lab Intelligent Percept & Image Understanding Minist Educ Xian 710071 Peoples R China Xidian Univ Acad Adv Interdisciplinary Res Xian 710071 Peoples R China Xidian Univ Sch Comp Sci & Technol Xian 710071 Peoples R China

As a unified framework for graph neural networks, message passing-based neural network (MPNN) has attracted a lot of research interest and has been shown successfully in a number of domains in recent years. However, because of over-smoothing and vanishing gradients, deep MPNNs are still difficult to train. To alleviate these issues, we first introduce a deep hierarchical layer aggregation (DHLA) strategy, which utilizes a block-based layer aggregation to aggregate representations from different layers and transfers the output of the previous block to the subsequent block, so that deeper MPNNs can be easily trained. Additionally, to stabilize the training process, we also develop a novel normalization strategy, neighbor normalization (NeighborNorm), which normalizes the neighbor of each node to further address the training issue in deep MPNNs. Our analysis reveals that NeighborNorm can smooth the gradient of the loss function, i.e., adding NeighborNorm makes the optimization landscape much easier to navigate. Experimental results on two typical graph pattern-recognition tasks, including node classification and graph classification, demonstrate the necessity and effectiveness of the proposed strategies for graph message-passing neural networks.

关键词： Training Analytical models data models Task analysis Smoothing methods Degradation Aggregates Deep graph neural networks graph data mining graph normalization graph representation learning

来源：评论

学校读者我要写书评

暂无评论

graph neural network with feature enhancement of isolated marginal groups

引用

APPLIED INTELLIGENCE 2022年第14期52卷 16962-16974页

作者： Wang, Jiahui Guo, Yi Wang, Zhihong Wen, Xinxiu Ni, Jianwei East China Univ Sci & Technol Shanghai Peoples R China Natl Engn Lab Big Data Distribut & Exchange Techn Shanghai Peoples R China Shanghai Engn Res Ctr Big Data & Internet Audienc Shanghai Peoples R China Minist Publ Secur Res Inst 3 Shanghai Peoples R China

graph neural networks (GNN) have achieved remarkable success by combining feature and structure information. However, the over-smoothing phenomenon has always been a crucial issue in GNN models since the node representation will easily converge to the full graph representation with the increasing of convolutional layers. Our investigation also indicates that the isolated part significantly restricts the capability due to the meaningless structure and local over-smoothing problem. While current models treated all nodes as equal with the absence of effectiveness of structure at different locations. To facilitate this line of research, we innovatively propose the graph neural network with feature enhancement of isolated parts (GNN-FEIP) consisting of graph partition, graph construction, and strategic label propagation procedures. In GNN-FEIP architecture, all the nodes are partitioned into several groups according to their position and connectivity. The feature-level similarity graph is reconstructed for subsequent feature enhancement of isolated nodes. Afterward, the preliminary prediction of the original GNN model has been adjusted with the strategic label propagation, which both balances the feature and structure of the nodes at different positions in a comprehensive manner. The exhaustive experiments indicate that GNN-FEIP achieves impressive performance than other off-the-shelf models, especially in the case that the isolated part is of large proportion.

关键词： graph neural network graph data mining Group partition Label propagation Isolated groups

来源：评论

学校读者我要写书评

暂无评论

A Network Learning Method for Functional Disability Prediction from Health data 16

A Network Learning Method for Functional Disability Predicti...

引用

16th International Conference on Knowledge Discovery and Information Retrieval, KDIR 2024 as part of 16th International Joint Conference on Knowledge Discovery, Knowledge Engineering and Knowledge Management, IC3K 2024

作者： Dondi, Riccardo Hosseinzadeh, Mehdi Università degli Studi di Bergamo Bergamo Italy University of Calabria CS Rende Italy

ISBN: (纸本)9789897587160

This contribution proposes a novel network analysis model with the goal of predicting a classification of individuals as either ‘disabled’ or ‘not-disabled’, using a dataset from the Health and Retirement Study (HRS). Our approach is based on selecting features that span health indicators and socioeconomic factors due to their pivotal roles in identifying disability. Considering the selected features, our approach computes similarities between individuals and uses this similarity to predict disability. We present a preliminary experimental evaluation of our method on the HRS dataset, where it shows an enhanced average accuracy of 62.48%. © 2024 by SCITEPRESS – Science and Technology Publications, Lda.

关键词： Disability Classification graph data mining Healthcare Analytics Learning Algorithms Network Analysis

来源：评论

学校读者我要写书评

暂无评论

CrsKGC-RSAN: Conflict Resolution Strategy for Knowledge graph Completion 23

CrsKGC-RSAN: Conflict Resolution Strategy for Knowledge Grap...

引用

23rd IEEE International Conference on data mining (IEEE ICDM)

作者： Chen, Jie Zhang, Xin Zhang, Wuyang Tan, Siyu Zhao, Shu Anhui Univ Sch Comp Sci & Technol Hefei Peoples R China

ISBN: (纸本)9798350381641

Knowledge graph Completion (KGC) is a crucial task aimed at enhancing the completeness and accuracy of knowledge graphs (KGs) by filling in missing information. However, existing knowledge graphs face two significant challenges: 1) the negative sampling issues, where false negative samples contain potential true facts, while easy negative samples limit the generalization capacity of KGC models;and 2) the entity conflict issues, many candidate entities exhibit similarities in their explicit structural features, making it difficult for models to make accurate decisions. To address the above challenges, we propose a novel method, CrsKGC-RSAN. For the negative sampling issues, we propose a Relative Self Adversarial Negative Sampling (RSAN) module. RSAN allows for a smaller margin between positive and negative samples compared to traditional selfadversarial sampling, while assigning higher training weights to hard negative samples to address false negative and esay negative sample issues. For the entity conflict issues, we present CrsKGC module, which, in addition to learning explicit structural features of KGs, mines implicit structural features, enhancing the comprehensiveness of feature learning and effectively resolving entity conflicts. Extensive experiments on the FB15K-237 and WN18RR datasets demonstrate that our method achieves stateof-the-art results and exhibits compatibility.

关键词： Knowledge graph Knowledge graph Completion Link Pediction Negative Sampling graph data mining

来源：评论

学校读者我要写书评

暂无评论

Propagation Measure on Circulation graphs for Tourism Behavior Analysis 22

Propagation Measure on Circulation Graphs for Tourism Behavi...

引用

37th Annual ACM Symposium on Applied Computing

作者： Prevoteau, Hugo Djebali, Sonia Zhao Laiping Travers, Nicolas Tianjin Univ Coll Intelligence & Comp Tianjin Peoples R China Leonard De Vinci Pole Univ Res Ctr F-92916 Paris France

ISBN: (纸本)9781450387132

Social network analysis has widespread in recent years, especially in digital tourism. Indeed the large amount of data that tourists produce during their travels represents an effective source to understand their behavior and is of great importance for tourism stakeholders. This paper studies the propagation effect of tourists on the territory thanks to geotagged circulation graphs. Those graphs reflect traffic flows which need to be analyzed over time and space. A new weighted measure is introduced for circulation characterization based on both topologies and distances. This measure helps to determine the behavior of tourists on local and global areas. An optimization strategy based on spanning trees is applied to reduce the computation on the whole graph while keeping a good approximation of the behavior. The approach is simulated on various graphs and evaluated experimentaly over a real dataset at various geographic zones, scales, communities, and time.

关键词： Spanning Trees graph data mining Digital Tourism

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：